Submit Search
Upload
HDFSvTACHYON
•
0 likes
•
179 views
Kevin Wong
Follow
Comparison between two distributed file systems: HDFS and Tachyon.
Read less
Read more
Software
Report
Share
Report
Share
1 of 18
Download now
Download to read offline
Recommended
Odoo Online platform: architecture and challenges
Odoo Online platform: architecture and challenges
Odoo
Some analysis of BlueStore and RocksDB
Some analysis of BlueStore and RocksDB
Xiao Yan Li
(SDD409) Amazon RDS for PostgreSQL Deep Dive | AWS re:Invent 2014
(SDD409) Amazon RDS for PostgreSQL Deep Dive | AWS re:Invent 2014
Amazon Web Services
Практический опыт профайлинга и оптимизации производительности Ruby-приложений
Практический опыт профайлинга и оптимизации производительности Ruby-приложений
Olga Lavrentieva
AWS RDS Benchmark - CMG Brasil 2012
AWS RDS Benchmark - CMG Brasil 2012
Rodrigo Campos
Amazon RDS for PostgreSQL - PGConf 2016
Amazon RDS for PostgreSQL - PGConf 2016
Grant McAlister
Ceph Day Santa Clara: Ceph Performance & Benchmarking
Ceph Day Santa Clara: Ceph Performance & Benchmarking
Ceph Community
Мониторинг. Опять, rootconf 2016
Мониторинг. Опять, rootconf 2016
Vsevolod Polyakov
Recommended
Odoo Online platform: architecture and challenges
Odoo Online platform: architecture and challenges
Odoo
Some analysis of BlueStore and RocksDB
Some analysis of BlueStore and RocksDB
Xiao Yan Li
(SDD409) Amazon RDS for PostgreSQL Deep Dive | AWS re:Invent 2014
(SDD409) Amazon RDS for PostgreSQL Deep Dive | AWS re:Invent 2014
Amazon Web Services
Практический опыт профайлинга и оптимизации производительности Ruby-приложений
Практический опыт профайлинга и оптимизации производительности Ruby-приложений
Olga Lavrentieva
AWS RDS Benchmark - CMG Brasil 2012
AWS RDS Benchmark - CMG Brasil 2012
Rodrigo Campos
Amazon RDS for PostgreSQL - PGConf 2016
Amazon RDS for PostgreSQL - PGConf 2016
Grant McAlister
Ceph Day Santa Clara: Ceph Performance & Benchmarking
Ceph Day Santa Clara: Ceph Performance & Benchmarking
Ceph Community
Мониторинг. Опять, rootconf 2016
Мониторинг. Опять, rootconf 2016
Vsevolod Polyakov
Metrics: where and how
Metrics: where and how
Vsevolod Polyakov
Путь мониторинга 2.0 всё стало другим / Всеволод Поляков (Grammarly)
Путь мониторинга 2.0 всё стало другим / Всеволод Поляков (Grammarly)
Ontico
Alluxio in MOMO
Alluxio in MOMO
Alluxio, Inc.
The Practice of Alluxio in Near Real-Time Data Platform at VIPShop [Chinese]
The Practice of Alluxio in Near Real-Time Data Platform at VIPShop [Chinese]
Alluxio, Inc.
Log
Log
Mariiana Guerrero
Aerospike & GCE (LSPE Talk)
Aerospike & GCE (LSPE Talk)
Sayyaparaju Sunil
Ceph at salesforce ceph day external presentation
Ceph at salesforce ceph day external presentation
Sameer Tiwari
Kauli SSPにおけるVyOSの導入事例
Kauli SSPにおけるVyOSの導入事例
Kazuhito Ohkawa
Stream Processing Inside Librato [Monitorama PDX 2015]
Stream Processing Inside Librato [Monitorama PDX 2015]
Librato, Inc.
Scaling with Python: SF Python Meetup, September 2017
Scaling with Python: SF Python Meetup, September 2017
Varun Varma
FASTER Key-Value Store and Log
FASTER Key-Value Store and Log
Badrish Chandramouli
"Metrics: Where and How", Vsevolod Polyakov
"Metrics: Where and How", Vsevolod Polyakov
Yulia Shcherbachova
Всеволод Поляков (DevOps Team Lead в Grammarly)
Всеволод Поляков (DevOps Team Lead в Grammarly)
Provectus
Partner Webinar: MongoDB and Softlayer on Bare Metal: Stability, Performance,...
Partner Webinar: MongoDB and Softlayer on Bare Metal: Stability, Performance,...
MongoDB
opentsdb in a real enviroment
opentsdb in a real enviroment
Chen Robert
Ceph Day Chicago - Supermicro Ceph - Open SolutionsDefined by Workload
Ceph Day Chicago - Supermicro Ceph - Open SolutionsDefined by Workload
Ceph Community
Gnocchi v4 (preview)
Gnocchi v4 (preview)
Gordon Chung
Pilot Hadoop Towards 2500 Nodes and Cluster Redundancy
Pilot Hadoop Towards 2500 Nodes and Cluster Redundancy
Stuart Pook
Gnocchi Profiling v2
Gnocchi Profiling v2
Gordon Chung
Bucket your partitions wisely - Cassandra summit 2016
Bucket your partitions wisely - Cassandra summit 2016
Markus Höfer
PacMin @ AMPLab All-Hands
PacMin @ AMPLab All-Hands
fnothaft
What’s New in the Berkeley Data Analytics Stack
What’s New in the Berkeley Data Analytics Stack
Turi, Inc.
More Related Content
What's hot
Metrics: where and how
Metrics: where and how
Vsevolod Polyakov
Путь мониторинга 2.0 всё стало другим / Всеволод Поляков (Grammarly)
Путь мониторинга 2.0 всё стало другим / Всеволод Поляков (Grammarly)
Ontico
Alluxio in MOMO
Alluxio in MOMO
Alluxio, Inc.
The Practice of Alluxio in Near Real-Time Data Platform at VIPShop [Chinese]
The Practice of Alluxio in Near Real-Time Data Platform at VIPShop [Chinese]
Alluxio, Inc.
Log
Log
Mariiana Guerrero
Aerospike & GCE (LSPE Talk)
Aerospike & GCE (LSPE Talk)
Sayyaparaju Sunil
Ceph at salesforce ceph day external presentation
Ceph at salesforce ceph day external presentation
Sameer Tiwari
Kauli SSPにおけるVyOSの導入事例
Kauli SSPにおけるVyOSの導入事例
Kazuhito Ohkawa
Stream Processing Inside Librato [Monitorama PDX 2015]
Stream Processing Inside Librato [Monitorama PDX 2015]
Librato, Inc.
Scaling with Python: SF Python Meetup, September 2017
Scaling with Python: SF Python Meetup, September 2017
Varun Varma
FASTER Key-Value Store and Log
FASTER Key-Value Store and Log
Badrish Chandramouli
"Metrics: Where and How", Vsevolod Polyakov
"Metrics: Where and How", Vsevolod Polyakov
Yulia Shcherbachova
Всеволод Поляков (DevOps Team Lead в Grammarly)
Всеволод Поляков (DevOps Team Lead в Grammarly)
Provectus
Partner Webinar: MongoDB and Softlayer on Bare Metal: Stability, Performance,...
Partner Webinar: MongoDB and Softlayer on Bare Metal: Stability, Performance,...
MongoDB
opentsdb in a real enviroment
opentsdb in a real enviroment
Chen Robert
Ceph Day Chicago - Supermicro Ceph - Open SolutionsDefined by Workload
Ceph Day Chicago - Supermicro Ceph - Open SolutionsDefined by Workload
Ceph Community
Gnocchi v4 (preview)
Gnocchi v4 (preview)
Gordon Chung
Pilot Hadoop Towards 2500 Nodes and Cluster Redundancy
Pilot Hadoop Towards 2500 Nodes and Cluster Redundancy
Stuart Pook
Gnocchi Profiling v2
Gnocchi Profiling v2
Gordon Chung
Bucket your partitions wisely - Cassandra summit 2016
Bucket your partitions wisely - Cassandra summit 2016
Markus Höfer
What's hot
(20)
Metrics: where and how
Metrics: where and how
Путь мониторинга 2.0 всё стало другим / Всеволод Поляков (Grammarly)
Путь мониторинга 2.0 всё стало другим / Всеволод Поляков (Grammarly)
Alluxio in MOMO
Alluxio in MOMO
The Practice of Alluxio in Near Real-Time Data Platform at VIPShop [Chinese]
The Practice of Alluxio in Near Real-Time Data Platform at VIPShop [Chinese]
Log
Log
Aerospike & GCE (LSPE Talk)
Aerospike & GCE (LSPE Talk)
Ceph at salesforce ceph day external presentation
Ceph at salesforce ceph day external presentation
Kauli SSPにおけるVyOSの導入事例
Kauli SSPにおけるVyOSの導入事例
Stream Processing Inside Librato [Monitorama PDX 2015]
Stream Processing Inside Librato [Monitorama PDX 2015]
Scaling with Python: SF Python Meetup, September 2017
Scaling with Python: SF Python Meetup, September 2017
FASTER Key-Value Store and Log
FASTER Key-Value Store and Log
"Metrics: Where and How", Vsevolod Polyakov
"Metrics: Where and How", Vsevolod Polyakov
Всеволод Поляков (DevOps Team Lead в Grammarly)
Всеволод Поляков (DevOps Team Lead в Grammarly)
Partner Webinar: MongoDB and Softlayer on Bare Metal: Stability, Performance,...
Partner Webinar: MongoDB and Softlayer on Bare Metal: Stability, Performance,...
opentsdb in a real enviroment
opentsdb in a real enviroment
Ceph Day Chicago - Supermicro Ceph - Open SolutionsDefined by Workload
Ceph Day Chicago - Supermicro Ceph - Open SolutionsDefined by Workload
Gnocchi v4 (preview)
Gnocchi v4 (preview)
Pilot Hadoop Towards 2500 Nodes and Cluster Redundancy
Pilot Hadoop Towards 2500 Nodes and Cluster Redundancy
Gnocchi Profiling v2
Gnocchi Profiling v2
Bucket your partitions wisely - Cassandra summit 2016
Bucket your partitions wisely - Cassandra summit 2016
Viewers also liked
PacMin @ AMPLab All-Hands
PacMin @ AMPLab All-Hands
fnothaft
What’s New in the Berkeley Data Analytics Stack
What’s New in the Berkeley Data Analytics Stack
Turi, Inc.
IndexedRDD: Efficeint Fine-Grained Updates for RDD's-(Ankur Dave, UC Berkeley)
IndexedRDD: Efficeint Fine-Grained Updates for RDD's-(Ankur Dave, UC Berkeley)
Spark Summit
A Survey of the State-of-the-art in Event Processing
A Survey of the State-of-the-art in Event Processing
Otávio Carvalho
Building Large Scale Machine Learning Applications with Pipelines-(Evan Spark...
Building Large Scale Machine Learning Applications with Pipelines-(Evan Spark...
Spark Summit
Alluxio Presentation at AMPLab Summer Retreat 2016
Alluxio Presentation at AMPLab Summer Retreat 2016
Alluxio, Inc.
AMP Camp 5 Intro
AMP Camp 5 Intro
jeykottalam
11. From Hadoop to Spark 1:2
11. From Hadoop to Spark 1:2
Fabio Fumarola
Alluxio (formerly Tachyon): The Journey thus far and the Road Ahead
Alluxio (formerly Tachyon): The Journey thus far and the Road Ahead
Alluxio, Inc.
The Next AMPLab: Real-Time, Intelligent, and Secure Computing
The Next AMPLab: Real-Time, Intelligent, and Secure Computing
Spark Summit
Viewers also liked
(10)
PacMin @ AMPLab All-Hands
PacMin @ AMPLab All-Hands
What’s New in the Berkeley Data Analytics Stack
What’s New in the Berkeley Data Analytics Stack
IndexedRDD: Efficeint Fine-Grained Updates for RDD's-(Ankur Dave, UC Berkeley)
IndexedRDD: Efficeint Fine-Grained Updates for RDD's-(Ankur Dave, UC Berkeley)
A Survey of the State-of-the-art in Event Processing
A Survey of the State-of-the-art in Event Processing
Building Large Scale Machine Learning Applications with Pipelines-(Evan Spark...
Building Large Scale Machine Learning Applications with Pipelines-(Evan Spark...
Alluxio Presentation at AMPLab Summer Retreat 2016
Alluxio Presentation at AMPLab Summer Retreat 2016
AMP Camp 5 Intro
AMP Camp 5 Intro
11. From Hadoop to Spark 1:2
11. From Hadoop to Spark 1:2
Alluxio (formerly Tachyon): The Journey thus far and the Road Ahead
Alluxio (formerly Tachyon): The Journey thus far and the Road Ahead
The Next AMPLab: Real-Time, Intelligent, and Secure Computing
The Next AMPLab: Real-Time, Intelligent, and Secure Computing
Similar to HDFSvTACHYON
Ycsb benchmarking
Ycsb benchmarking
Sqrrl
Cassandra TK 2014 - Large Nodes
Cassandra TK 2014 - Large Nodes
aaronmorton
System Interconnects for HPC
System Interconnects for HPC
inside-BigData.com
Flash for the Real World – Separate Hype from Reality
Flash for the Real World – Separate Hype from Reality
Hitachi Vantara
Intro to hadoop
Intro to hadoop
Haden Pereira
Технологии работы с дисковыми хранилищами и файловыми системами Windows Serve...
Технологии работы с дисковыми хранилищами и файловыми системами Windows Serve...
Виталий Стародубцев
Sql server scalability fundamentals
Sql server scalability fundamentals
Chris Adkin
HBaseCon 2015: HBase 2.0 and Beyond Panel
HBaseCon 2015: HBase 2.0 and Beyond Panel
HBaseCon
Compaction, Compaction Everywhere
Compaction, Compaction Everywhere
DataStax Academy
Nosql series-part-3-hypertable
Nosql series-part-3-hypertable
hypertable
FAQ on Dedupe NetApp
FAQ on Dedupe NetApp
Ashwin Pawar
Keynote Hadoop Summit Dublin 2016: Hadoop Platform Innovations - Pushing The ...
Keynote Hadoop Summit Dublin 2016: Hadoop Platform Innovations - Pushing The ...
Sumeet Singh
Exchange 2010 storage improvements
Exchange 2010 storage improvements
Nathan Winters
Demystifying Storage
Demystifying Storage
bhavintu79
Sql server 2016 it just runs faster sql bits 2017 edition
Sql server 2016 it just runs faster sql bits 2017 edition
Bob Ward
Demystifying Storage - Building large SANs
Demystifying Storage - Building large SANs
Directi Group
ORC 2015: Faster, Better, Smaller
ORC 2015: Faster, Better, Smaller
DataWorks Summit
Accumulo Summit 2015: Ferrari on a Bumpy Road: Shock Absorbers to Smooth Out ...
Accumulo Summit 2015: Ferrari on a Bumpy Road: Shock Absorbers to Smooth Out ...
Accumulo Summit
Need For Speed- Using Flash Storage to optimise performance and reduce costs-...
Need For Speed- Using Flash Storage to optimise performance and reduce costs-...
NetAppUK
(DAT402) Amazon RDS PostgreSQL:Lessons Learned & New Features
(DAT402) Amazon RDS PostgreSQL:Lessons Learned & New Features
Amazon Web Services
Similar to HDFSvTACHYON
(20)
Ycsb benchmarking
Ycsb benchmarking
Cassandra TK 2014 - Large Nodes
Cassandra TK 2014 - Large Nodes
System Interconnects for HPC
System Interconnects for HPC
Flash for the Real World – Separate Hype from Reality
Flash for the Real World – Separate Hype from Reality
Intro to hadoop
Intro to hadoop
Технологии работы с дисковыми хранилищами и файловыми системами Windows Serve...
Технологии работы с дисковыми хранилищами и файловыми системами Windows Serve...
Sql server scalability fundamentals
Sql server scalability fundamentals
HBaseCon 2015: HBase 2.0 and Beyond Panel
HBaseCon 2015: HBase 2.0 and Beyond Panel
Compaction, Compaction Everywhere
Compaction, Compaction Everywhere
Nosql series-part-3-hypertable
Nosql series-part-3-hypertable
FAQ on Dedupe NetApp
FAQ on Dedupe NetApp
Keynote Hadoop Summit Dublin 2016: Hadoop Platform Innovations - Pushing The ...
Keynote Hadoop Summit Dublin 2016: Hadoop Platform Innovations - Pushing The ...
Exchange 2010 storage improvements
Exchange 2010 storage improvements
Demystifying Storage
Demystifying Storage
Sql server 2016 it just runs faster sql bits 2017 edition
Sql server 2016 it just runs faster sql bits 2017 edition
Demystifying Storage - Building large SANs
Demystifying Storage - Building large SANs
ORC 2015: Faster, Better, Smaller
ORC 2015: Faster, Better, Smaller
Accumulo Summit 2015: Ferrari on a Bumpy Road: Shock Absorbers to Smooth Out ...
Accumulo Summit 2015: Ferrari on a Bumpy Road: Shock Absorbers to Smooth Out ...
Need For Speed- Using Flash Storage to optimise performance and reduce costs-...
Need For Speed- Using Flash Storage to optimise performance and reduce costs-...
(DAT402) Amazon RDS PostgreSQL:Lessons Learned & New Features
(DAT402) Amazon RDS PostgreSQL:Lessons Learned & New Features
Recently uploaded
Call Girls In Mukherjee Nagar 📱 9999965857 🤩 Delhi 🫦 HOT AND SEXY VVIP 🍎 SE...
Call Girls In Mukherjee Nagar 📱 9999965857 🤩 Delhi 🫦 HOT AND SEXY VVIP 🍎 SE...
Call Girls In Delhi Whatsup 9873940964 Enjoy Unlimited Pleasure
(Genuine) Escort Service Lucknow | Starting ₹,5K To @25k with A/C 🧑🏽❤️🧑🏻 89...
(Genuine) Escort Service Lucknow | Starting ₹,5K To @25k with A/C 🧑🏽❤️🧑🏻 89...
gurkirankumar98700
CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online ☂️
CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online ☂️
anilsa9823
Adobe Marketo Engage Deep Dives: Using Webhooks to Transfer Data
Adobe Marketo Engage Deep Dives: Using Webhooks to Transfer Data
BradBedford3
Der Spagat zwischen BIAS und FAIRNESS (2024)
Der Spagat zwischen BIAS und FAIRNESS (2024)
OPEN KNOWLEDGE GmbH
A Secure and Reliable Document Management System is Essential.docx
A Secure and Reliable Document Management System is Essential.docx
ComplianceQuest1
Russian Call Girls in Karol Bagh Aasnvi ➡️ 8264348440 💋📞 Independent Escort S...
Russian Call Girls in Karol Bagh Aasnvi ➡️ 8264348440 💋📞 Independent Escort S...
soniya singh
TECUNIQUE: Success Stories: IT Service provider
TECUNIQUE: Success Stories: IT Service provider
mohitmore19
Building a General PDE Solving Framework with Symbolic-Numeric Scientific Mac...
Building a General PDE Solving Framework with Symbolic-Numeric Scientific Mac...
stazi3110
Unveiling the Tech Salsa of LAMs with Janus in Real-Time Applications
Unveiling the Tech Salsa of LAMs with Janus in Real-Time Applications
Alberto González Trastoy
Hand gesture recognition PROJECT PPT.pptx
Hand gesture recognition PROJECT PPT.pptx
bodapatigopi8531
5 Signs You Need a Fashion PLM Software.pdf
5 Signs You Need a Fashion PLM Software.pdf
Wave PLM
How To Use Server-Side Rendering with Nuxt.js
How To Use Server-Side Rendering with Nuxt.js
Andolasoft Inc
Exploring iOS App Development: Simplifying the Process
Exploring iOS App Development: Simplifying the Process
Evangelist Apps https://twitter.com/EvangelistSW/
Diamond Application Development Crafting Solutions with Precision
Diamond Application Development Crafting Solutions with Precision
SolGuruz
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
OnePlan Solutions
Software Quality Assurance Interview Questions
Software Quality Assurance Interview Questions
Arshad QA
The Ultimate Test Automation Guide_ Best Practices and Tips.pdf
The Ultimate Test Automation Guide_ Best Practices and Tips.pdf
kalichargn70th171
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
ICS
Advancing Engineering with AI through the Next Generation of Strategic Projec...
Advancing Engineering with AI through the Next Generation of Strategic Projec...
OnePlan Solutions
Recently uploaded
(20)
Call Girls In Mukherjee Nagar 📱 9999965857 🤩 Delhi 🫦 HOT AND SEXY VVIP 🍎 SE...
Call Girls In Mukherjee Nagar 📱 9999965857 🤩 Delhi 🫦 HOT AND SEXY VVIP 🍎 SE...
(Genuine) Escort Service Lucknow | Starting ₹,5K To @25k with A/C 🧑🏽❤️🧑🏻 89...
(Genuine) Escort Service Lucknow | Starting ₹,5K To @25k with A/C 🧑🏽❤️🧑🏻 89...
CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online ☂️
CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online ☂️
Adobe Marketo Engage Deep Dives: Using Webhooks to Transfer Data
Adobe Marketo Engage Deep Dives: Using Webhooks to Transfer Data
Der Spagat zwischen BIAS und FAIRNESS (2024)
Der Spagat zwischen BIAS und FAIRNESS (2024)
A Secure and Reliable Document Management System is Essential.docx
A Secure and Reliable Document Management System is Essential.docx
Russian Call Girls in Karol Bagh Aasnvi ➡️ 8264348440 💋📞 Independent Escort S...
Russian Call Girls in Karol Bagh Aasnvi ➡️ 8264348440 💋📞 Independent Escort S...
TECUNIQUE: Success Stories: IT Service provider
TECUNIQUE: Success Stories: IT Service provider
Building a General PDE Solving Framework with Symbolic-Numeric Scientific Mac...
Building a General PDE Solving Framework with Symbolic-Numeric Scientific Mac...
Unveiling the Tech Salsa of LAMs with Janus in Real-Time Applications
Unveiling the Tech Salsa of LAMs with Janus in Real-Time Applications
Hand gesture recognition PROJECT PPT.pptx
Hand gesture recognition PROJECT PPT.pptx
5 Signs You Need a Fashion PLM Software.pdf
5 Signs You Need a Fashion PLM Software.pdf
How To Use Server-Side Rendering with Nuxt.js
How To Use Server-Side Rendering with Nuxt.js
Exploring iOS App Development: Simplifying the Process
Exploring iOS App Development: Simplifying the Process
Diamond Application Development Crafting Solutions with Precision
Diamond Application Development Crafting Solutions with Precision
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Software Quality Assurance Interview Questions
Software Quality Assurance Interview Questions
The Ultimate Test Automation Guide_ Best Practices and Tips.pdf
The Ultimate Test Automation Guide_ Best Practices and Tips.pdf
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
Advancing Engineering with AI through the Next Generation of Strategic Projec...
Advancing Engineering with AI through the Next Generation of Strategic Projec...
HDFSvTACHYON
1.
HDFS v TachyonKevin
Wong
2.
110x faster 2x
faster
3.
AMPLab’s Machines r3.8xlarge 244GB 32
cores
4.
My Machines m1.large 7.5GB 2
cores
5.
Word Count on
4-Node ClusterInputSize(GB) 1 3 5 10 30 50 100 300 500 Throughput (GB/sec) 0 0.008 0.015 0.023 0.03 HDFS Tachyon Broke! Broke!
6.
Word Count on
10-Node ClusterInputSize(GB) 1 3 5 10 30 50 100 300 Throughput (GB/sec) 0 0.01 0.02 0.03 0.04 HDFS Tachyon
7.
Word Count on
20-Node ClusterInputSize(GB) 1 3 5 10 30 50 100 300 Throughput (GB/sec) 0 0.015 0.03 0.045 0.06 HDFS Tachyon
8.
Word Count on
25-Node ClusterInputSize(GB) 1 3 5 10 30 50 100 300 Throughput (GB/sec) 0 0.018 0.035 0.053 0.07 HDFS Tachyon
9.
SQL Join on
25-Node ClusterInputSize(GB) 1 3 5 10 30 50 100 300 Throughput (GB/sec) 0 0.125 0.25 0.375 0.5 HDFS Tachyon
10.
Challenges • Could not
use MapReduce, so had to use Spark • Scala and python • Files don’t fit in memory
11.
Read and Write 70
seconds Tachyon HDFS 54 seconds
12.
Configurations --executor-memory 3G --executor-memory 4G --executor-memory
5G
13.
5x r3.8xlarge 244GB Tachyon: 200GB Spark:
20GB
14.
Read and Write 22
seconds Tachyon HDFS 2354 seconds 107x faster
15.
No change • Word
Count • SQL Join • Reads
16.
For moderate machines,
don’t use Tachyon. Use Tachyon on machines with large memory running write-heavy jobs.
Download now